sign in sign up
<<
Home , Synaptic vesicle, Synaptotagmin, Vesicle fusion

LASKER MEDICAL BASIC RESEARCH AWARD ESSAY

A molecular machine for release: and beyond

Thomas C Südhof

Fifty years ago, ’s seminal work revealed that calcium triggers neurotransmitter membrane release by stimulating ultrafast synaptic . But how a presynaptic terminal achieves the speed and precision of calcium-triggered fusion remained unknown. My colleagues and I set out to study this fundamental problem more C than two decades ago. Munc18 How do the synaptic vesicle and the plasma membrane fuse during transmitter release? How does calcium trigger synaptic vesicle fusion? Synaptotagmin? How is calcium influx localized to release sites N in order to enable the fast coupling of an action α-, β- or γ-SNAP– potential to transmitter release? Together with NSF complex? contributions made by other scientists, most prominently James Rothman, Reinhard Jahn and SNAP-25 (HPC-1) Richard Scheller, and assisted by luck and good fortune, we have addressed these questions over the last decades. Presynaptic plasma membrane © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature As I describe below, we now know of a general mechanism of membrane fusion that Figure 1 Description of the SNARE–SM complex that mediates synaptic vesicle fusion. The initial operates by the interaction of SNAREs (for sol- study2 identified the SM protein Munc18-1 as a component of the fusion machinery that also contains npg uble N-ethylmaleimide–sensitive factor (NSF)- the SNARE synaptobrevin (also known as VAMP), SNAP-25 and syntaxin (also known as HPC-1), attachment protein receptors) and SM proteins although the detailed protein-protein interactions involved were defined only later (reproduced from ref. 2). (for Sec1/Munc18-like proteins). We also have now a general mechanism of calcium-triggered Starting my lab cells, but the molecular mechanisms involved fusion that operates by calcium binding to syn- When I started my laboratory in 1986 at the in these fusion reactions were unclear. The lack aptotagmins, plus a general mechanism of ves- University of Texas Southwestern in Dallas, of knowledge about how synaptic vesicle fusion icle positioning adjacent to calcium channels, exquisite electrophysiological studies had already happens and how such fusion is controlled by which involves the interaction of the so-called characterized neurotransmitter release in detail. calcium intrigued me and led me to search for RIM proteins with these channels and synap- These studies showed that calcium triggers molecular mechanisms. tic vesicles. Thus, a molecular framework that release within a few hundred microseconds, We chose a simple approach: to isolate and accounts for the astounding speed and preci- exhibits amazing plasticity and displays a non- clone all of the major proteins present in pre- sion of neurotransmitter release has emerged. linear dependence on calcium. However, aside synaptic terminals. Initially, in collaboration with In describing this framework, I have been asked from calcium, not a single molecule important Reinhard Jahn, we focused on synaptic vesicles to describe primarily my own work. I apologize for release had been identified. because they could be isolated at high yield and for the many omissions of citations to work of Genetic screens by Sidney Brenner, Randy purity. Later on, we expanded this approach to others; please consult a recent review for addi- Scheckman and their colleagues had isolated the presynaptic and plasma mem- tional references1. gene mutations that disrupt synaptic transmis- brane. The goal was to achieve a molecular cata- sion in or impair the log of protein components of the presynaptic Thomas C. Südhof is at the Department of secretory pathway in yeast, but the function of terminal as a starting point for a functional Molecular and Cellular Physiology and the Howard the corresponding proteins were unknown. In dissection. For over a decade, we purified and Hughes Medical Institute, Stanford University pioneering work, Rothman performed in vitro cloned a series of major synaptic proteins— School of Medicine, Palo Alto, California, USA. membrane fusion assays using non-neuronal among others, , synaptobrevin,

NATURE MEDICINE VOLUME 19 | NUMBER 10 | OCTOBER 2013 1227 ESSAY

Membranes poised for fusion the action of SM proteins; see below) opens the Presynaptic R-SNARE fusion pore (step 3). Fusion-pore expansion terminal (synaptobrevin/ VAMP) 4. Snare complex disassembly SV 1. SNARE complex assembly transforms the initial trans-SNARE complexes and vesicle recycling (chaperones: CSPα, β, or γ + into cis-SNARE complexes that are then disso- Qa-SNARE SM protein (chaperones: NSF + (syntaxin) Qb- & Qc-SNARE (Munc18) α-, β- or γ-synucleins) ciated by the ATPase NSF, which binds SNARE α-, β- or γ-SNAPs) (SNAP-25) complexes via a-, b- or g-SNAP adaptor proteins, N thereby completing the cycle (step 4). In a physiological context, SNARE complex Synaptic cleft assembly forces opposing membranes into Membranes after fusion Membranes primed for fusion close proximity but is insufficient to mediate cis-SNARE complex– Partial trans-SNARE complex– SM protein assembly SM protein assembly fusion. Our initial proposal that SM proteins SNARE– generally contribute to membrane fusion2 SM protein N N met with skepticism because biochemical N N cycle data seemed to argue against this hypothesis. However, studies over the past decade have revealed that Munc18-1 remains associated Fusion pore opening during fusion with SNARE proteins through- trans-SNARE complex/ SM protein assembly out their assembly-disassembly cycle and that 4 3. Fusion pore 2. Fusion pore this association is essential for fusion . This finding resolved the questions surrounding expansion N N opening Munc18-1 and strongly supported the idea that Munc18-1 and other SM proteins are obligatory Fusion pore components of fusion machines1. Figure 2 Model of the SNARE–SM protein cycle during synaptic vesicle fusion. During step 1, synaptic vesicles are primed for fusion; this step involves opening of the closed conformation of syntaxin, a How calcium controls membrane fusion switch of the Munc18-binding mode of syntaxin from the closed to the open conformation and partial The above discussion describes the major prog- assembly of trans-SNARE complexes. Step 1 is facilitated by recently discovered chaperones (cysteine ress that was made in determining the mecha- string proteins (CSPs) and synucleins) that enhance SNARE complex assembly and whose dysfunction nism of membrane fusion. At the same time, my 3 is related to neurodegeneration . During step 2, the fusion pore opens, with full trans-SNARE complex laboratory was focusing on a question crucial for assembly. During step 3, the fusion pore expands, converting trans-SNARE into cis-SNARE complexes. neuronal function: how is this process triggered In step 4, NSF and SNAPs mediate disassembly of the SNARE complex, leading to vesicle recycling. The cycle shown here for synaptic vesicle fusion is paradigmatic for most cytoplasmic fusion reactions, in microseconds when calcium enters the pre- although the details differ. In knockout experiments, deletion of the SM protein Munc18-1 produces synaptic terminal? the most severe phenotype22, possibly because loss of SNARE components of the fusion machinery is While examining the fusion machinery, we better compensated for than loss of SM proteins. SNAREs are generally classified into four types wondered how it could possibly be controlled so (R, Qa, Qb and Qc) that assemble into SNARE complexes in an obligatory R-Qa-Qb-Qc combination. tightly by calcium. Starting with the description of synaptotagmin-1 (Syt1)5, we worked over two © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature , , Munc13s, Munc18s, activities of botulinum and toxins, which decades to show that calcium-dependent exocy- , RIMs, RIM-BPs, and were known to block neurotransmitter release. tosis is mediated by synaptotagmins as calcium neuroligins. These findings strongly supported the idea that sensors. npg the three SNARE proteins function in synaptic Synaptotagmins are evolutionarily conserved Elucidating the mechanism of fusion vesicle fusion. Thus, at the end of 1993, most transmembrane proteins with two cytoplasmic 5,6 As a molecular catalog of synaptic vesicle pro- components of the synaptic fusion machinery C2 domains (Fig. 3a) . When we cloned Syt1, teins emerged, the question of how synaptic had been identified. But how do SNARE and nothing was known about C2 domains except vesicles may interact with the presynaptic plasma SM proteins mediate fusion? It took two decades that they represented the ‘second constant membrane became approachable. Other labora- until a plausible model emerged1. sequence’ in protein- C isozymes. Because tories cloned SNAP-25 and syntaxin-1, presyn- During fusion, SNARE and SM proteins had been shown to interact with aptic plasma membrane proteins now known as undergo a cycle of association and dissociation, by an unknown mechanism, we t-SNAREs. In a crucial study, Rothman showed which is maintained by chaperones that sup- speculated that Syt1 C2 domains may bind phos- that syntaxin and SNAP-25 form a complex port SNARE complex assembly (cysteine string pholipids, which we indeed found to be the case5. (which he named the SNARE complex) with proteins and synucleins)3 and disassembly (NSF We also found that this interaction is calcium 6,7 the synaptic vesicle protein that we had identi- and SNAPs; Fig. 2). The underlying principle dependent and that a single medi- fied and named synaptobrevin (also cloned by of SNARE and SM protein function is simple: ates calcium-dependent binding 8 Scheller, who named it VAMP). Immediately SNARE proteins are attached to both mem- (Fig. 3b) . In addition, the Syt1 C2 domains also afterwards, we identified Munc18-1 as a syn- branes destined to fuse and form a trans com- bind syntaxin-1 and the SNARE complex6,9. All taxin-associated protein and proposed that plex that involves a progressive zippering of the of these observations were first made for Syt1 C2 Munc18-1 and the SNARE proteins constitute four-helical SNARE complex bundle in an N- to domains, but they have since been generalized to 2 the fusion machine of synaptic vesicles (Fig. 1) . C-terminal direction (step 1 in Fig. 2). The zip- other C2 domains. In parallel, the laboratories of Cesare pering of trans-SNARE complexes forces the two As calcium-binding modules, C2 domains Montecucco and Jahn—partly in collaboration fusing membranes into close proximity, destabi- were unlike any other calcium-binding pro- with our lab—identified synaptobrevin, SNAP- lizing the membrane surfaces (step 2). Assembly tein known at the time. Beginning in 1995, we 25 and syntaxin-1 as substrates for the proteolytic of the full trans-SNARE complex (together with obtained atomic structures of calcium-free and

1228 VOLUME 19 | NUMBER 10 | OCTOBER 2013 NATURE MEDICINE ESSAY

a 3 Ca2+ 2 Ca2+ Figure 3 Discovery of synaptotagmins and of the precision of release by association with SNARE calcium-binding properties of C2 domains. (a) Syt1 complexes and phospholipids (Fig. 6a,b). N- -C cloning defined the canonical domain structure It was initially surprising that the Syt1 knock- C2AC2B of synaptotagmins, which includes a single out produced a marked phenotype because the SV membrane 4 transmembrane region and two C2 domains . brain expresses multiple synaptotagmins6. Only some synaptotagmins are glycosylated in However, we found that only three synaptotag- b 25 their intravesicular sequence as shown; some other synaptotagmins bind calcium in a different mins—Syt1, Syt2 and Syt9—mediate fast syn- ) 13 –3 20 stoichiometry or not at all. SV, synaptic vesicle. aptic vesicle . Syt2 triggers release (b) Calcium-dependent phospholipid binding by faster, and Syt9 slower, than Syt1. Most fore- C A 2 the Syt1 C2A domain. Note the steep calcium brain express only Syt1, but not Syt2 or 15 domain cooperativity and micromolar calcium affinity of Syt9, accounting for the profound Syt1 knock- binding (modified from ref. 7). Data shown are out phenotype. Syt2 is the predominant calcium means ± s.e.m.; y axis depicts the amount of 10 14 binding to the immobilized Syt1 C2A sensor of very fast in the brainstem , domain measured as the amount of tritiated whereas Syt9 is primarily present in the limbic 3 13 H-PC bound (c.p.m. x 10 H-PC bound (c.p.m. 5 Control phosphotidylcholine ( H-PC) bound. (c) Atomic system . Thus, the kinetic properties of Syt1, 3 structures of the Syt1 C2A and C2B domains with Syt2 and Syt9 correspond to the functional three and two bound calcium ions, respectively. 0 needs of the synapses that contain them. –7 –6 –5 –4 Images show ribbon diagrams with cyan-colored 10 10 10 10 Parallel experiments in neuroendocrine cells Free Ca2+ concentration (M) b-strands, orange a-helices and orange balls for the calcium ions (courtesy of J. Rizo, University of revealed that, in addition to Syt1, Syt7 functions C A C B Texas Southwestern). as a calcium sensor for exocytosis. c 2 2 domain domain Moreover, experiments in olfactory neurons with calcium binding to Syt1 performing a role uncovered a role for Syt10 as a calcium sensor unrelated to calcium sensing and transmitter for insulin-like growth factor-1 exocytosis15, release. To directly test whether calcium bind- showing that, even in a single , differ- ing to Syt1 triggers release, we introduced a ent synaptotagmins act as calcium sensors for point mutation into the endogenous mouse distinct fusion reactions. Viewed together with Syt1 gene locus. This mutation decreased the results by other groups, these observations indi- Syt1 calcium-binding affinity by about twofold11. cated that calcium-triggered exocytosis generally Electrophysiological recordings revealed that depends on synaptotagmin calcium sensors and this mutation also decreased the calcium affinity that different synaptotagmins confer specificity of neurotransmitter release approximately two- onto exocytosis pathways. fold, formally proving that Syt1 is the calcium We had originally identified as sensor for release (Fig. 5). In addition to medi- a small protein bound to SNARE complexes ating calcium triggering of release, Syt1 controls (Fig. 6b)16. Analysis of complexin-deficient (‘clamps’) the rate of spontaneous release occur- neurons showed that complexin represents calcium-bound Syt1 C2 domains10 in collabo- ring in the absence of action potentials, thus a cofactor for synaptotagmin that functions © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature ration with structural biologists, primarily Jose serving as an essential mediator of the speed and both as a clamp and as an activator of calcium- Rizo (Fig. 3c). These structures provided the first insights into how C2 domains bind calcium and ab npg allowed us to test the role of Syt1 calcium binding in transmitter release11. The biochemical properties of Syt1 sug- gested that it constituted Katz’s long-sought calcium sensor for neurotransmitter release. Initial experiments in C. elegans and , 160 pA however, disappointingly indicated otherwise. 50 ms The ‘synaptotagmin calcium-sensor hypothesis’ seemed unlikely until our electrophysiological analyses of Syt1 knockout mice revealed that Syt1 is required for all fast synchronous synap- 160 pA tic fusion in forebrain neurons but is dispens- 50 ms able for other types of fusion (Fig. 4)12. These experiments established that Syt1 is essential for fast calcium-triggered release, but not for fusion as such. Although the Syt1 knockout analysis sup- 40 pA 40 pA ported the synaptotagmin calcium-sensor 50 ms 50 ms hypothesis, it did not exclude the possibility Figure 4 Syt1 knockout ablates fast synchronous transmitter release. (a,b) Traces showing recordings from that Syt1 positions vesicles next to voltage-gated neurons cultured from littermate wild-type (a) and Syt1-knockout (b) mice. Traces depict synaptic responses calcium channels (a function now known to be to isolated action potentials; standard (top) and expanded views (bottom) illustrate that the Syt1 knockout mediated by RIMs and RIM-BPs; see below), completely ablates fast synchronous release but not slow asynchronous responses (modified from ref. 12).

NATURE MEDICINE VOLUME 19 | NUMBER 10 | OCTOBER 2013 1229 ESSAY

+ NH ab3 H237 Ca2+ (μM) c D238 2.0 – WT L171 Loop 1 M 0.1 2.0 3.0 4.0 8.0 250 r Input R233Q K236Q Ca2 Ca3 53K 1. 5 S235 M173 R233Q D172 35K H – – – 1. 0 F234 G174 53K 0.5 D232 F231 Ca1 G175 WT 35K

Loop 3 Normalized amplitude NH D178 T176 0 R233Q – 110 C 2+ + S177 Ca (mM) H2N NH2

Figure 5 Introducing a mutation that changes the calcium affinity of Syt1 into endogenous Syt1 in mice establishes that Syt1 is a calcium sensor for neurotransmitter release. (a) Calcium-binding site architecture of the Syt1 C2A domain and location of the substitution (R233Q) that we analyzed. A second innocuous mutation (K236Q) served as a control. (b) The R233Q substitution decreases the apparent calcium affinity of Syt1 during phospholipid binding approximately twofold. Data show binding of a native Syt1 fragment containing both C2 domains to liposomes as a function of the free calcium concentration. Mr, apparent molecular mass. (c) The R233Q mutation decreases the apparent calcium affinity of synaptic exocytosis by about twofold. Figure depicts amplitudes of synaptic responses induced at different concentrations of extracellular calcium (modified from ref. 11). WT, wild type. Data show means ± s.e.m.; lines are fitted by a Hill function with an approximately twofold lower apparent calcium affinity for R233Q than for WT neurons.

triggered fusion17. Complexin-deficient neu- and that at least part of complexin competes recruitment, vesicle rons exhibit a phenotype milder than that of with synaptotagmin for SNARE complex bind- docking and priming Syt1-deficient neurons, with a selective sup- ing. Calcium-activated synaptotagmin displaces Up to this point, I have focused on two of Katz’s pression of fast synchronous exocytosis and this part of complexin, thereby triggering fusion- central questions—how membranes fuse and an increase in spontaneous exocytosis, which pore opening (Fig. 6a)1,18. how such fusion is controlled by calcium. But suggests that complexin and synaptotagmins a are functionally interdependent. Synaptobrevin/ How does a small molecule like complexin, Synaptotagmin-1 SV VAMP composed of only ~130 amino acid residues, act NSF-mediated Assembly of SNARE complex prefusion SNARE– to activate and clamp synaptic vesicles for synap- disassembly & SM protein complex vesicle recycling totagmin action? Atomic structures revealed that, Munc18 when bound to assembled SNARE complexes, Munc13 Docking complexin contains two short a-helices flanked SNAP-25 Syntaxin-1 RIMs ADP+ a P & by flexible sequences (Fig. 6c). One of the -heli- i ces is bound to the SNARE complex and is essen- NSF SNAPs tial for all complexin function18. The second

© 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature Fully assembled Partly asssembled a-helix is required only for the clamping, and cis-SNARE complex trans-SNARE complex not for the activating function of complexin17. The flexible N-terminal sequence of complexin, npg conversely, mediates only the activating, but not the clamping, function of the protein. Our cur- Fusion completion Priming I rent model is that complexin binding to SNAREs NSF SNAPs Complexin Activation of activates the SNARE–SM protein complex Fusion pore ATP & prefusion expansion + SNARE/SM Ca2+-binding sites NSF & SNAP protein binding 2+ complex by Figure 6 Model of the action of synaptotagmin and Ca complexin

complexin in the SNARE–SM protein cycle. 2+ (a) Overview of the SNARE–SM protein cycle Ca and the points of action of complexin and synaptotagmin. Superimposed on the SNARE– Fusion pore opening Priming II SM protein cycle (Fig. 2) is the association of Ca2+-triggering fusion pore opening complexin with partially assembled trans-SNARE complexes, which enhances vesicle priming for Ca2+ Complexin fusion, and the step of calcium triggering of fusion bc Central a-helix Accessory a-helix pore opening by synaptotagmin. (b) Expanded (SNARE binding) (clamping) view of the primed synaptic vesicle fusion complex Synaptotagmin N N-terminal Synapto-brevin/ C-terminal (an assembly of SNARE proteins, Munc18-1 C sequence VAMP sequence (activating) and complexin) with synaptotagmin poised to Ca2+ Munc18 trigger fusion pore opening upon calcium binding. N (c) Modular domain structure of complexin C with functional assignments of the complexin sequences, and a space-filling model of the atomic SNAP-25 Syntaxin structure of the SNARE complex containing bound Complexin complexin (modified from ref. 17).

1230 VOLUME 19 | NUMBER 10 | OCTOBER 2013 NATURE MEDICINE ESSAY

V.P. Whittaker and J.L. Goldstein for continuous advice, Cytosol Synaptic and my co-workers and collaborators for invaluable x vesicle guidance and support, in particular R. Jahn, R.E. Hammer and J. Rizo. I am grateful to the Howard Hughes Medical Institute and the NIMH and NINDS, Rab3/Rab27 US National Institutes of Health, for financial support. C2 x Synaptotagmin-1 Munc13 GTP COMPETING FINANCIAL INTERESTS The author declares no competing financial interests.

x Synaptobrevin/ 1. Südhof, T.C. & Rothman, J.E. Membrane fusion: grap- C2 VAMP pling with SNARE and SM proteins. Science 323, 474– RIM C 477 (2009). x 2 2. Hata, Y., Slaughter, C.A. & Südhof, T.C. Synaptic vesicle Munc18 fusion complex contains unc-18 homologue bound to Ca2+ syntaxin. Nature 366, 347–351 (1993). RIM-BP 3. Burré, J. et al. a-synuclein promotes SNARE-complex assembly in vivo and in vitro. Science 329, 1663–1667 SNAP-25 (2010). Complexin 4. Khvotchev, M. et al. Dual modes of Munc18–1/SNARE interactions are coupled by functionally critical bind- ing to syntaxin-1 N-terminus. J. Neurosci. 27, 12147– 12155 (2007). 2+ Ca channel Syntaxin 5. Perin, M.S., Fried, V.A., Mignery, G.A., Jahn, R. & Ca2+ Südhof, T.C. Phospholipid binding by a synaptic vesicle Synaptic cleft protein homologous to the regulatory region of protein kinase C. Nature 345, 260–263 (1990). Figure 7 Diagram of the protein complex that mediates the recruitment of calcium channels and the 6. Li, C. et al. Ca2+-dependent and Ca2+-independent activ- docking of vesicles at release sites. RIM, RIM-BP and Munc13 are multidomain proteins that form a ities of neural and nonneural synaptotagmins. Nature tight complex that mediates three essential functions of active zones: recruitment of Ca2+ channels to 375, 594–599 (1995). 7. Brose, N., Petrenko, A.G., Südhof, T.C. & Jahn, R. enable tight coupling of action potentials to release by localizing calcium influx next to the calcium Synaptotagmin: a Ca2+ sensor on the synaptic vesicle sensor synaptotagmin, docking of vesicles at the release site and Munc13-dependent priming of surface. Science 256, 1021–1025 (1992). the fusion machinery. Spheres denote calcium ions; of the domains shown, only calcium-binding C2 8. Davletov, B.A. & Südhof, T.C. A single C2-domain from domains are specifically labeled (modified from ref. 20). synaptotagmin I is sufficient for high affinity Ca2+/ phospholipid-binding. J. Biol. Chem. 268, 26386– 26390 (1993). I have not addressed the third question that is calcium channels and a loss of the tight coupling 9. Pang, Z.P., Shin, O.-H., Meyer, A.C., Rosenmund, C. crucial to understand the speed of synaptic trans- of a presynaptic to release. & Südhof, T.C. A gain-of-function mutation in synap- totagmin-1 reveals a critical role of Ca2+-dependent mission: how is calcium influx localized to the RIMs perform their functions in a direct SNARE-complex binding in synaptic exocytosis. active zone for rapid coupling of an action poten- complex with the calcium channels, with other J. Neurosci. 26, 12556–12565 (2006). tial to neurotransmitter release? Only recently we active zone proteins such as RIM-BPs (which, in 10. Sutton, R.B., Davletov, B.A., Berghuis, A.M., Südhof, T.C. & Sprang, S.R. Structure of the first C2-domain of syn- identified a molecular mechanism that addresses turn, also bind calcium channels) and with Rab3/ aptotagmin I: a novel Ca2+/phospholipid binding fold. this question and ties everything together. Rab27 GTPases on synaptic vesicles (Fig. 3). 80, 929–938 (1995). 11. Fernández-Chacón, R. et al. Synaptotagmin I functions A molecular mechanism that explains how At the same time, we and others found that as a Ca2+-regulator of release probability. Nature 410, calcium channels and synaptic vesicles are RIMs are essential for normal vesicle dock- 41–49 (2001).

© 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature 2+ localized to the active zone emerged with the ing at the active zone, thus tying docking and 12. Geppert, M. et al. Synaptotagmin I: a major Ca sen- sor for transmitter release at a central . Cell 79, demonstration that two families of active-zone calcium channel recruitment together (Fig. 7). 717–727 (1994). proteins, RIMs and RIM-BPs, collaborate to The role of RIMs and RIM-BPs in recruiting 13. Xu, J., Mashimo, T. & Südhof, T.C. Synaptotagmin-1, -2, and -9: Ca2+-sensors for fast release that specify distinct npg 19,20 recruit calcium channels to release sites . calcium channels and docking vesicles to active presynaptic properties in subsets of neurons. Neuron 54, The same proteins also dock synaptic vesicles zones is evolutionarily conserved and represents 567–581 (2007). 2+ at release sites and bind Munc13 proteins, which a fundamental mechanism underlying synaptic 14. Sun, J. et al. A dual Ca -sensor model for neuro- transmitter release in a central synapse. Nature 450, catalyze the priming of SNARE–SM protein transmission. 676–682 (2007). complexes for fusion19–21. Thus, in a parsimo- 15. Cao, P., Maximov, A. & Südhof, T.C. Activity-dependent Outlook IGF-1 exocytosis is controlled by the Ca2+-sensor nious design, a single protein complex localizes synaptotagmin-10. Cell 145, 300–311 (2011). synaptic vesicles and calcium channels next to Our work, together with that of other research- 16. McMahon, H.T., Missler, M., Li, C. & Südhof, T.C. release sites at the active zone and recruits the ers, uncovered a plausible mechanism explaining Complexins: cytosolic proteins that regulate SNAP- receptor function. Cell 83, 111–119 (1995). priming factor Munc13 to the vesicles (Fig. 7). how membranes undergo rapid fusion during 17. Maximov, A., Tang, J., Yang, X., Pang, Z. & Südhof, T.C. Our initial studies identified RIMs, RIM-BPs transmitter release, how such fusion is regulated Complexin controls the force transfer from SNARE com- plexes to membranes in fusion. Science 323, 516–521 and Munc13 proteins as multidomain scaffolding by calcium and how the calcium-controlled (2009). proteins of the presynaptic active zone that form fusion of synaptic vesicles is spatially organized 18. Tang, J. et al. Complexin/synaptotagmin-1 switch con- in the presynaptic terminal. Nevertheless, many trols fast synaptic vesicle exocytosis. Cell 126, 1175– a complex with each other, prime vesicles for 1187 (2006). fusion and mediate short-term synaptic plastic- new questions now arise that are not just details 19. Wang, Y., Okamoto, M., Schmitz, F., Hofman, K. & ity19–21. Identification of the role of these proteins but of great importance. For example, what are Südhof, T.C. RIM: a putative Rab3-effector in regulating synaptic vesicle fusion. Nature 388, 593–598 (1997). in recruiting calcium channels to the active zone, the precise physicochemical mechanisms under- 20. Kaeser, P.S. et al. RIM proteins tether Ca2+-channels to however, was only possible when we deleted all lying fusion, and what is the role of the fusion presynaptic active zones via a direct PDZ-domain inter- RIM isoforms from presynaptic terminals19. mechanism we outlined in brain diseases? Much action. Cell 144, 282–295 (2011). 21. Schoch, S. et al. RIM1a forms a protein scaffold for We found that RIMs selectively bind calcium remains to be done in this field. regulating neurotransmitter release at the active zone. channels expressed in presynaptic active zones Nature 415, 321–326 (2002). 22. Verhage, M. et al. Synaptic assembly of the brain in and that deletion of RIMs causes a decrease of ACKNOWLEDGMENTS the absence of neurotransmitter . Science 287, presynaptic calcium influx, a loss of presynaptic I thank my lifelong mentors in science M.S. Brown, 864–869 (2000).

NATURE MEDICINE VOLUME 19 | NUMBER 10 | OCTOBER 2013 1231